My last post touches on this. The Psamtik intercepts the Belisarius after the Belisarius makes a clearly vectored jump, and it doesn't even have its weapons completely ready when it arrives. The Ravana jumps in after a rolling engagement that lasted quite a while. The Colossus arrived at a pre-planned ambush point. The Beast and Sathanas #?? jump the Colossus after a lengthy engagement with Shivans on the scene.
Jumps are stochastic. Subspace isn't simply connected to realspace. Depending on the configuration of mass in the system, the exact point you're departing from, the recent use history of your drive and power systems, your computational resources, the amount of hurry you're in, the precision of your observers on the far end of the jump, and a bunch of peculiar subspace factors, it may be easy to show up in a good firing position — or you may only get a satisfactory arrival solution a long ways out.
(...)
BP is based on the fiction of retail FS2, where the Shivans do not repeatedly drop their destroyers in exactly optimal position.
Shivans don't use their capabilities optimally (because of the nature of how they think and their ultimate goals).
I don't think FS canon specifies in any point that precise jumps are easier to calculate if other ships were deployed in the area recently. It is possible and sounds realistic, but for my best knowledge it's non-canon. No ship behaviour or techroom entry in FS2 indicates that. Where the hell in FS2 are Shivans jumping in as far as Shivan capships in AoA?
Having beams is an excellent reason to use stand-off engagements against opponents who can't hit back at that range.
Exactly. And if, as a Shivan, you want to use this advantage, you have to arrive WITHIN fire range, in other case you have to hold enemy bombers until you can fire and your advantage may likely go to hell.
The Vigilant was ambushed at the Gamma Draconis jump node, which is one of two strategically interesting places in the entire system.
I know. I used this example to point out Shivans don't need time to adapt to new enemy's tactics as you suggested about the Demon.
So, let's sum up. If we assume (non-canonically, but it's ok, as you developed the FS2 lore much and added lots of new elements) that:
1) The larger the ship is, the less precise the jump will be
2) If other ships have perfomed jumps to a location recently, it is easier to calculate exact coordinates for next ones
3) There are plenty of factors difficult to foreseen that can make calculating jump coordinates more difficult
And refer this to AoA missions, then we can draw some conclusions:
1) Frankenstein's Monsters, Demon-attacking-Duke - mission can be explained with first contact with GTVA and no Shivans in the area, although the Demon could easily deploy fighters first to do the reckon and then calc more precise jump.
2) Preserving the Balance, with the three destroyers jumping in in a distance - no explanation you gave suits here. Temeraire was in a defensive position, it was not moving at all, Shivans were launching assaults against her including two destroyers and bombers, they knew their exact location. They could easily jump closer.
3) Time For Heroes - size of Sathanas and no Shivans in region is a kind of explanation here. Although basing GTVA tactics on a simple fact that this time the enemy will jump far enough so he have time to destroy the beams is risky.